DYNATRACE TRAINING
TOKEN: dt0c01.DBGNW5K6QISR355R42NPYQLB.F443TRNSURCBG2BFQ5Y
DYNATRACE FEATURES:
Observability AI feature:
Full stack monitoring
Problem analysis
Cloud Monitoring: Integration with cloud watch tool
Client side, browser monitoring, user experience monitoring, Real user monitoring and synthetic
[robotic] monitoring [tool will perform certain task via job].
Exposes Data using REST API
BROWSER LEVEL, WEB SERVICES LEVEL, VALIDATE SEQUENCE PAGE NAVIGATION
[DIGITAL] SYNTHETIC MONITORING: MAIN FOCUS IS CHECKING AVAILABILITY + PERFORMANCE CHECK [
CONCEPTS IN OUTSIDE WORLD:
smart scape
Distributed traces – Pure path
One agent
Memory snapshot / dump – heap dump [.hprof extension] – Java visual Vm OR Heaphero.io website for
analysis.
Thread snapshot – thread dump [ Not in Dynatrace]
Memory dump analysis is done to understand inefficient object and wasted memory
Garbage collection summary for freeup – top 20 objects
Classes – by instance percentage
Instances
https://www.oracle.com/webfolder/technetwork/tutorials/obe/java/gc01/index.html
Decision Tree:
Problems Notification
Dashboards
Smartscape Topology
Distributed Traces (Purepath)
----------------------------------------------------------------------
Real Monitoring /Synthetic Monitoring
Services
Database
Host
Network
DYNATRACE TRAINING
Data lake: unstructured and raw data.
Data Warehouse: structured or semi-structured data
Dynatrace active gate works as a proxy between on agent and DT server, it collects ad aggregates
monitoring data, secures it through encrypted https connection and send it to DT server, it require full
internet access.
It can be also used for VMWare, DT Synthetics and RUM.
DEM RUM is about customer journey and DT Synthetics across global network
DYNATRACE TRAINING
It can be integrated with Amazon web services, for intelligent monitoring of services running amazon
cloud,
PAAS environment monitoring like cloud foundry and Microsoft azure web apps.
VMWare – DT shows how vms affect webapps, on can use RUM for agentless monitoring by ingesting
Java script tags on each web pages [ only when deep monitoring cannot be used on web servers] to
measure app performance.
Mobile Monitoring - DT provides a wizard that lets you package to ship with app.
Synthetic : Public and password protected web urls, ur level uptime and performance.
DT App Engine – Allows building of custom apps by inheriting centralized app engine logic.
Automation Engine – Has task that can be used to achieve efficiency and innovate.
a. Monaco: Automation as code, a configuration as code tool.
Dynatrace Platform subscription: flexible access: annual commit [on demand consumption] & Rate card
Account Management: myaccount.dynatrace.com
License Overview: total usage by date, env and tenant
Lens: Adoption: Specific user activity , Env: Technology used
Notification: Event Type: consumption / forecast
Goals: Redundancy, Security & Always ON [Availability]
Metric and Transaction data is encrypted during transit and at rest.
Security Certification: AICPA/Soc 2/SAS 70, ISO27001
Masking Strategy: Anonymization API & Data masking
DYNATRACE TRAINING
Kubernetes: Utilization, workload, events and vulnerabilities.
Amazon Elastic Kub services
Google Kub service
Amazon/Azure
Cloud foundry: Gorouters [Traffic], Diego cells [run containers], Auctioneers [actions].
Dockers/Containers: Container analysis, process groups, events and logs
Apache Mesos: Requires Paas token, separate app.
Network Monitoring: If overhead increases me than 5% of available CPU then DT pauses Network
module by 3 min max up to 45 min.
Automated discovery with consistent identity management and metadata propagation is most crucial
for maintaining monitoring continuity in ephemeral environments. This architectural component
enables Dynatrace to instantly detect new infrastructure components as they appear, automatically
instrument them for monitoring, and maintain consistent entity identity even as specific instances come
and go. The metadata propagation ensures that monitoring context, tags, and configurations
automatically apply to new instances based on their characteristics rather than requiring manual
assignment. This dynamic approach maintains continuous monitoring in environments where instances
may live for just minutes or hours and where the scale might change dramatically within seconds –
conditions that would make manual configuration, static IP assignments, or periodic rescanning entirely
ineffective for maintaining consistent observability.
Memory analysis with object allocation hotspots and class histogram evolution is most valuable for
identifying memory leaks in long-running Java applications. This sophisticated capability goes beyond
basic heap monitoring by analyzing the actual objects being created, tracking which code locations
allocate the most memory, and monitoring how the distribution of object types (class histogram)
changes over time. By identifying classes whose instance counts continuously increase or code locations
that allocate excessive objects, this analysis can detect memory leaks during their early stages – long
before they consume enough memory to trigger garbage collection problems or outages. This proactive
detection enables addressing memory leaks before they impact production, providing insights that basic
heap metrics, garbage collection tracking, or thread monitoring cannot deliver with the same precision.
